Dictionary of Evidence-based Medicine.pdf

More documents

Recommendations

Info

Dictionary of Evidence-based Medicine 13 times it lands heads, we would expect this to be close to 0.5. If we repeat this experiment many times and ignore the occasions when the proportion is exactly 0.5, on average half of the time, the proportions will be on the low side of 0.5 and the other half on the high side. If the coin were loaded, then we would expect the two proportions to be unequal to an extent which will be proportional to the loading. In clinical studies, two common sources of bias are (i) poor sampling leading to non-representative samples being assessed (e.g. estimating the efficacy of a smoking cessation aid in a general population using inpatient post-myocardial patients as a sampling frame) and (ii) flaw in the measurement method or process (e.g. use of a faulty sphygmomanometer in blood pressure measurement). Information bias occurs when systematic differences are introduced in the measurement of the response. A well-known example is recall bias in case-control studies. Patients (cases) with, say, an adverse reaction are more likely to recall exposure to a suspected drug than are control subjects. Non-response bias refers to bias which may be introduced when those who respond may differ systematically from non-responders. For this reason, those conducting surveys take great care to design their studies to minimize non-response and to make adjustments in their analyses to account for any non-response. Protopathic bias (also referred to as reverse causality) is the term applied when a bias arises from misclassification, for example when general awareness of the study being carried out leads to exposure after the index date being reported more frequently in the case patients than in the controls. Selection bias arises in comparative studies when the investigator inadvertently assigns patients with different characteristics which have a bearing on the outcome to the various groups. An often quoted example is that of a trial of a travel sickness remedy. Passengers were allocated to the control group and the ship's crew to the treatment group (the captain thought he could not afford to have travel-sick crew!). Staging bias is a type of selection bias in which patients with different levels of disease severity are assigned in a biased way to the different treatment groups. This causes confounding by disease severity. In a review of comparative studies of radical prostatectomy and irradiation for the management of patients with early-stage prostate cancer, the authors found that patients who received radiation therapy tended to have more aggressive tumours than those who received surgery. Moreover, in the surgery group, surgery may be abandoned if examination of the regional lymph nodes early in the procedure reveals evidence of cancer spread. By expressing results in terms of survival for patients without spread at the time of surgery, surgical patients were bound to perform better than those
14 Dictionary of Evidence-based Medicine subjected to radiation, a group which included patients with disseminated disease (Wasson J, Cushman C, Bruskewitz R et al. (1993) A structured literature review of treatment for localised prostate cancer. Archives of Family Medicine. 2: 487-93). Binary contingent valuation (see under Contingent valuation) Binary variable A variable for which there are only two outcomes (e.g. living status - dead or alive - or outcome of tossing a coin). Binomial distribution Consider a series of Bernoulli trials (i.e. each trial has only two outcomes). Suppose that the probability of success in each trial is the same and equal to p and that n trials are undertaken. If Y is the random variable representing the number of successes in the n trials, then Y follows the binomial probability distribution given by: where y is an actual observation and (\ = (l-p). An example which is easy to relate to is a series of tosses of the same coin. Each toss of the coin is a Bernoulli trial with probability of obtaining a head of 0.5 for a fair coin. y is the number of heads in the n trials. Note that 0! is defined as 1. The binomial distribution has mean np and variance npq. Binomial theorem The binomial theorem gives the formula for expanding expressions of the form (x + y} n :
Page 2 and 3: Dictionary of Evidence-based Medici
Page 4 and 5: Preface This book was conceived at
Page 6 and 7: A Ability to pay Ability to pay is
Page 12 and 13: Figure I Audit: Clinical audit cycl
Page 14 and 15: B Balanced design In an experiment,
Page 66 and 67: Figure 9 Concept of externality Dic
Page 68 and 69:
Dictionary of Evidence-based Medici
Page 70 and 71:
Page 72 and 73:
G Game theory Game theory, first de
Page 74 and 75:
Page 76 and 77:
Page 78 and 79:
Page 80 and 81:
Page 82 and 83:
Page 84 and 85:
Page 86 and 87:
I ICD (see International Classifica
Page 88 and 89:
Page 90 and 91:
Page 92 and 93:
Page 94 and 95:
K Keynes, John Maynard (1883-1946)
Page 96 and 97:
Page 98 and 99:
Page 100 and 101:
M Macroeconomics Macroeconomics is
Page 102 and 103:
Page 104 and 105:
Page 106 and 107:
Page 108 and 109:
Page 110 and 111:
Page 112 and 113:
N N of I trial An N of 1 trial is a
Page 114 and 115:
Page 116 and 117:
o OBM (see Opinion-based medicine)
Page 118 and 119:
Page 120 and 121:
Page 122 and 123:
Page 124 and 125:
Page 126 and 127:
Page 128 and 129:
Page 130 and 131:
Page 132 and 133:
Page 134 and 135:
Page 136 and 137:
Page 138 and 139:
Page 140 and 141:
Figure 25 Measuring quality of life
Page 142 and 143:
Page 144 and 145:
Page 146 and 147:
Page 148 and 149:
Page 150 and 151:
S Scarcity Scarcity is often referr
Page 152 and 153:
Page 154 and 155:
Page 156 and 157:
Page 158 and 159:
T TDM (see Therapeutic drug monitor
Page 160 and 161:
Page 162 and 163:
u Unique value effect This term is
Page 164 and 165:
Page 166 and 167:
Page 168 and 169:
Page 170:
Z Z variate A random variable which
show all

Dictionary of Evidence-based Medicine.pdf

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?